A selection interview lasts anywhere from a few minutes to half an hour. During that time, the person sitting across from me does exactly what they should do, and they are right to do it: they present themselves at their best, they have prepared their answers, they know which questions are coming. I call this the seduction mask. There is nothing dishonest about it: we all wear masks, every day, and change them with the time of day and the circumstances. It is part of the exercise, and a candidate unable to wear it would be rather more worrying.
That is not the problem. The problem is that the mask holds for the length of the interview without effort, and that it tells me almost nothing I am trying to learn. I am not trying to find out whether someone presents their project well. I am trying to find out how they work when no one is watching, how they react when contradicted, and what they do with a day that does not go as planned. Starting a business exposes an entrepreneur to uncertainty, to setbacks, and to other people: staff, clients, suppliers, bankers, partners. The posture is not the one you adopt in front of a program’s “recruiter”.
This is why I had added a final stage to the selection. We brought the shortlisted candidates together for one or two days, ran a dense program, and I stayed in observation, with colleagues. Among ourselves, we called that moment the washing machine1.
What duration brings out
The first thing that appears is the gap between what a person declares and what they do on site. Someone can spend an hour describing their working method, their discipline, the way they organise themselves. Over two days, you see their working method during individual and group exercises.
The second thing is the response to contradiction. When you challenge a candidate’s assumption during an interview, they have time to answer politely and move on. When you challenge it on the first day, they have to live with that contradiction until the next morning. Some return with something they have built, because they worked on it overnight. Others defend their original position with more insistence than the day before. That difference appears in no application file. It is what determines coachability.
The third thing is what happens when the frame opens. At some point in the program, there is no precise instruction. Some people create their own task. Others wait to be told what to do, and the wait can last a long time.
An expensive instrument
I removed that stage after two cohorts, for a simple reason: it required time I no longer had. Two days of continuous presence, with real attention paid to fifteen or twenty people, sometimes more, cannot easily be delegated and cannot be compressed. The quality of what I observed came precisely from my being there, with resource persons2, from beginning to end.
I do not regret it, because removing a component is part of design work. An instrument that produces excellent information but that its designer can no longer run will end up producing mediocre information. Better to retire it cleanly than to let it degrade.
What I keep from it is a conviction about method. Observation time is a design variable in the same way selection criteria are. When you shorten it, you do not only reduce the cost of the program, you change the nature of what the selection is able to see. That choice deserves to be made deliberately rather than by default.
- The term was popularised by Anglo-Saxon accelerators, following the Y Combinator or Techstars model, to describe express qualification sessions, often a bootcamp of two to five days, in which candidates have to prove their worth in practice. ↩
- A resource person is an experienced professional whose role is to observe and give feedback, a comment or a suggestion on what is happening in the room. I discovered the practice during my certification as a trainer, a few years ago. ↩